Loss of heterozygosity (LOH) in pediatric and adult tumors indicates that chromosome band 11p15.5 harbors one or more growth or tumor suppressor genes. This notion is supported by growth arrest and tumor suppression studies using RD and G401 cell hybrid functional assays. In addition, the genes responsible for Beckwith-Wiedemann syndrome (BWS; an overgrowth and cancer predisposition disorder) and Long QT syndrome map to this region. The applicants have isolated this important region in PAC clones generating a 1l0-1.1 mb contig between D11S601 and IGF2/H19. They have located nine known genes in this contig and identified 18 novel transcripts. Tissue specific expression patterns of these novel genes have been determined by northern blotting. Since this region is imprinted, allele-specific expression is being assessed by conventional methods as well as a novel somatic cell hybrid assay. Two novel genes (one of which is imprinted) are overlapping and divergently transcribed, and exhibit their highest level of expression in fetal and adult liver and kidney making them candidates for tumor suppressors in hepatoblastoma and Wilms' tumor. Three BWS rearrangement breakpoints and a rhabdoid tumor breakpoint have been shown to disrupt the Long QT (KVLQT1) gene. The applicants also show that one of these rearrangements is associated with relaxation of genomic imprinting at IGF2 and recognition of a novel differentially methylated CpG-island. Studies are proposed to characterize 11p15.5 novel genes with respect to their expression in tumors. Those genes exhibiting an appropriate expression profile will be screened for mutations in Wilms' tumor, rhabdomyosarcomas and breast and ovarian carcinomas The tumor suppressor potential of candidate genes will be tested in a functional assay by expressing them in RD and G401 cells. Studies are also proposed to identify epigenetic changes in tumors and BWS patients as well as to identify a 11p15 imprinting cancer (IC). These studies will further our understanding of the molecular pathology of cancer including the involvement of genomic imprinting. This information may identify valuable prognostic markers and will facilitate more rational approaches to cancer therapies.